Re: [PATCH] gnu: Add sendmail

[Alex Kost]

Alex Vong (2016-09-24 16:22 +0800) wrote:

> Alex Kost  writes:
>
>> John Darrington (2016-09-17 12:11 +0200) wrote:
>>
>>> On Sat, Sep 17, 2016 at 05:38:26PM +0800, Alex Vong wrote:
>> [...]
>>>  >  > + "contrib/mmuegel" "devtools/bin/configure.sh")
>>>  >  > +   (("/bin/sh") (which "bash")))
>>>  >  > +
>>>  >  > + (substitute* "devtools/bin/Build"
>>>  >  > + (("SHELL=/bin/sh") (string-append "SHELL=" (which "bash"
>>>  >  > + #t))
>>>  >  I think the `#t' is not neccessary here, since `substitute*' 
>>> uses
>>>  >  `substitute', which will either return #t or throw an exception.
>>>  >
>>>  > WTF??  Didn't you complain earlier this week when I *didn't* put #t 
>>> in
>>>  > exactly this
>>>  > scenario??
>>>  >
>>>  Yes, I am a different Alex :)
>>>  Also, it seems we are not being consistent here, sometimes we put `#t'
>>>  after `substitute*', sometimes we don't. Anyone has an idea?
>>>
>>> I did raise some suggestions in my earlier posts.  But again I don't
>>> have any strong
>>> opinion.
>>
>> I have a strong opinion: if a docstring of a procedure says what value
>> it returns, we can rely on it, otherwise we should not guess what value
>> will be returned.  In case of 'substitute*' (and 'substitute'), the
>> returned value is not specified, so I think if a phase ends with
>> 'substitute*', we should (or even must) add #t after it.
>
> I see your point that one should not be relying on undocumented
> features, which I agree. But I also see an alternative: to make
> 'substitute*' either return true or throw an exception and document
> it. I think the heart of the problem is scheme is "untyped", so we rely
> on the documentation. What do you think?

I don't think my opinion matters, I'm not the man to decide.  As for me,
I would prefer if 'substitute*' returned #t if substitution was
successful and #f if it failed (if nothing was substituted).

OTOH 'substitute*' is only one of the several procedures that are used
often in phases; there also can be met such general guile procedures as
'setenv', 'copy-file', etc.  After them (in the end of the phases) we
have to use #t anyway.

-- 
Alex

Re: [PATCH] gnu: Add sendmail

[Alex Vong]

Alex Kost  writes:

> John Darrington (2016-09-17 12:11 +0200) wrote:
>
>> On Sat, Sep 17, 2016 at 05:38:26PM +0800, Alex Vong wrote:
> [...]
>>  >  > + "contrib/mmuegel" "devtools/bin/configure.sh")
>>  >  > +   (("/bin/sh") (which "bash")))
>>  >  > +
>>  >  > + (substitute* "devtools/bin/Build"
>>  >  > + (("SHELL=/bin/sh") (string-append "SHELL=" (which "bash"
>>  >  > + #t))
>>  >  I think the `#t' is not neccessary here, since `substitute*' uses
>>  >  `substitute', which will either return #t or throw an exception.
>>  >
>>  > WTF??  Didn't you complain earlier this week when I *didn't* put #t in
>>  > exactly this
>>  > scenario??
>>  >
>>  Yes, I am a different Alex :)
>>  Also, it seems we are not being consistent here, sometimes we put `#t'
>>  after `substitute*', sometimes we don't. Anyone has an idea?
>>
>> I did raise some suggestions in my earlier posts.  But again I don't
>> have any strong
>> opinion.
>
> I have a strong opinion: if a docstring of a procedure says what value
> it returns, we can rely on it, otherwise we should not guess what value
> will be returned.  In case of 'substitute*' (and 'substitute'), the
> returned value is not specified, so I think if a phase ends with
> 'substitute*', we should (or even must) add #t after it.

I see your point that one should not be relying on undocumented
features, which I agree. But I also see an alternative: to make
'substitute*' either return true or throw an exception and document
it. I think the heart of the problem is scheme is "untyped", so we rely
on the documentation. What do you think?

Re: [PATCH] gnu: Add sendmail

[Alex Kost]

John Darrington (2016-09-17 12:11 +0200) wrote:

> On Sat, Sep 17, 2016 at 05:38:26PM +0800, Alex Vong wrote:
[...]
>  >  > +"contrib/mmuegel" 
> "devtools/bin/configure.sh")
>  >  > +   (("/bin/sh") (which "bash")))
>  >  > +
>  >  > + (substitute* "devtools/bin/Build"
>  >  > + (("SHELL=/bin/sh") (string-append "SHELL=" (which "bash"
>  >  > + #t))
>  >  I think the `#t' is not neccessary here, since `substitute*' uses
>  >  `substitute', which will either return #t or throw an exception.
>  >
>  > WTF??  Didn't you complain earlier this week when I *didn't* put #t in
>  > exactly this
>  > scenario??
>  >
>  Yes, I am a different Alex :)
>  Also, it seems we are not being consistent here, sometimes we put `#t'
>  after `substitute*', sometimes we don't. Anyone has an idea?
>
> I did raise some suggestions in my earlier posts.  But again I don't have any 
> strong
> opinion.

I have a strong opinion: if a docstring of a procedure says what value
it returns, we can rely on it, otherwise we should not guess what value
will be returned.  In case of 'substitute*' (and 'substitute'), the
returned value is not specified, so I think if a phase ends with
'substitute*', we should (or even must) add #t after it.

-- 
Alex

Re: [PATCH] gnu: Add sendmail

[John Darrington]

On Sat, Sep 17, 2016 at 05:38:26PM +0800, Alex Vong wrote:
 Hi,
 
 John Darrington  writes:
 
 > They need more than that.  They need the binary to be installed
 > setuid.  So these
 > cannot be run as package tests.  They must be tested after the
 > as-yet-to-be-written
 > service is installed.  But I'll update the comment to make it more clear.
 >
 I see, so this is a circular dependency problem here. The service
 depends on this package, but the test depends on the service. Am I
 right?

More or less :)  
 
 >  > +   #:phases
 >  > +   (modify-phases %standard-phases
 >  > + (add-before 'build 'replace-/bin/sh
 >  > +   (lambda _
 >  > + (substitute*
 >  > + (append
 >  > + (list "smrsh/smrsh.c" "sendmail/conf.c" "contrib/mailprio"
 >  > +"contrib/mmuegel" 
"devtools/bin/configure.sh")
 >  > +  (find-files "." ".*\\.m4")
 >  > +  (find-files "." ".*\\.cf"))
 >  I think this can be simplified to:
 >  `("smrsh/smrsh.c" "sendmail/conf.c"
 > "contrib/mailprio" "contrib/mmuegel"
 > "devtools/bin/configure.sh"
 > .@(find-files "." ".*\\.m4")
 > ,@(find-files "." ".*\\.cf"))
 >  using the quasi-quote quasi-unquote-splicing notation, which is 
similar
 >  to string interpolation in shell "foo bar $(CAR) $(TAR)".
 >
 > Is that simpler?  It has more characters?
 >
 Hmmm, it is more of a style thing. You can ask how others think about
 it.

I don't have any strong opinions on either style.
 
 >  > +"contrib/mmuegel" 
"devtools/bin/configure.sh")
 >  > +   (("/bin/sh") (which "bash")))
 >  > +
 >  > + (substitute* "devtools/bin/Build"
 >  > + (("SHELL=/bin/sh") (string-append "SHELL=" (which "bash"
 >  > + #t))
 >  I think the `#t' is not neccessary here, since `substitute*' uses
 >  `substitute', which will either return #t or throw an exception.
 >
 > WTF??  Didn't you complain earlier this week when I *didn't* put #t in
 > exactly this
 > scenario??
 >
 Yes, I am a different Alex :)
 Also, it seems we are not being consistent here, sometimes we put `#t'
 after `substitute*', sometimes we don't. Anyone has an idea?

I did raise some suggestions in my earlier posts.  But again I don't have any 
strong
opinion.
 
 > Would that make any difference? Since only the last expression would
 > get returned from lambda ?
 >  
 You are right. The following should chain things up correctly:
 
 (and (zero? (system* "sh" "Build"))
  (with-directory-excursion "cf/cf"
(begin
  (copy-file "generic-linux.mc" "sendmail.mc")
  (zero? (system* "sh" "Build" "sendmail.cf")

You're right.  I'll do it that way.
 

Thanks,

J'

-- 
Avoid eavesdropping.  Send strong encrypted email.
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.



signature.asc
Description: Digital signature

Re: [PATCH] gnu: Add sendmail

[Alex Vong]

Hi,

John Darrington  writes:

> They need more than that.  They need the binary to be installed
> setuid.  So these
> cannot be run as package tests.  They must be tested after the
> as-yet-to-be-written
> service is installed.  But I'll update the comment to make it more clear.
>
I see, so this is a circular dependency problem here. The service
depends on this package, but the test depends on the service. Am I
right?

>  > +   #:phases
>  > +   (modify-phases %standard-phases
>  > + (add-before 'build 'replace-/bin/sh
>  > +   (lambda _
>  > + (substitute*
>  > + (append
>  > + (list "smrsh/smrsh.c" "sendmail/conf.c" "contrib/mailprio"
>  > +"contrib/mmuegel" "devtools/bin/configure.sh")
>  > +  (find-files "." ".*\\.m4")
>  > +  (find-files "." ".*\\.cf"))
>  I think this can be simplified to:
>  `("smrsh/smrsh.c" "sendmail/conf.c"
> "contrib/mailprio" "contrib/mmuegel"
> "devtools/bin/configure.sh"
> .@(find-files "." ".*\\.m4")
> ,@(find-files "." ".*\\.cf"))
>  using the quasi-quote quasi-unquote-splicing notation, which is similar
>  to string interpolation in shell "foo bar $(CAR) $(TAR)".
>
> Is that simpler?  It has more characters?
>
Hmmm, it is more of a style thing. You can ask how others think about
it.

>  > +"contrib/mmuegel" "devtools/bin/configure.sh")
>  > +   (("/bin/sh") (which "bash")))
>  > +
>  > + (substitute* "devtools/bin/Build"
>  > + (("SHELL=/bin/sh") (string-append "SHELL=" (which "bash"
>  > + #t))
>  I think the `#t' is not neccessary here, since `substitute*' uses
>  `substitute', which will either return #t or throw an exception.
>
> WTF??  Didn't you complain earlier this week when I *didn't* put #t in
> exactly this
> scenario??
>
Yes, I am a different Alex :)
Also, it seems we are not being consistent here, sometimes we put `#t'
after `substitute*', sometimes we don't. Anyone has an idea?

>  > + (replace 'configure
>  > +   (lambda _
>  > +
>  > + ;; Render harmless any attempts to chown or chmod
>  > + (substitute* "devtools/bin/install.sh"
>  > +   (("owner=\\$2") "owner=''")
>  > +   (("group=\\$2") "group=''"))
>  > +
>  > + (with-output-to-file "devtools/Site/site.config.m4"
>  > +   (lambda ()
>  > + (format #t "
>  > +define(`confCC', `gcc')
>  > +define(`confOPTIMIZE', `-g -O2')
>  > +define(`confLIBS', `-lresolv')
>  > +define(`confINSTALL', `~a/devtools/bin/install.sh')
>  > +define(`confDEPEND_TYPE', `CC-M')
>  > +define(`confINST_DEP', `')
>  > +" (getcwd))
>  > + (replace 'build
>  > +   (lambda _
>  > + (system* "sh" "Build")
>  I think there is a missing `zero?' here.
>
>
> Would that make any difference? Since only the last expression would
> get returned from lambda ?
>  
You are right. The following should chain things up correctly:

(and (zero? (system* "sh" "Build"))
 (with-directory-excursion "cf/cf"
   (begin
 (copy-file "generic-linux.mc" "sendmail.mc")
 (zero? (system* "sh" "Build" "sendmail.cf")

>  > + (with-directory-excursion "cf/cf"
>  > +   (begin
>  > + (copy-file "generic-linux.mc" "sendmail.mc")
>  > + (zero? (system* "sh" "Build" "sendmail.cf"))
>  > + (add-before 'install 'pre-install
>  > +   (lambda _
>  > + (let ((out (assoc-ref %outputs "out")))
>  > +   (mkdir-p (string-append out "/usr/bin"))
>  > +   (mkdir-p (string-append out "/usr/sbin"))
>  > +   (mkdir-p (string-append out "/etc/mail"))
>  > +   (setenv "DESTDIR" out)
>  > +   (with-directory-excursion "cf/cf"
>  > + (zero? (system* "sh" "Build" "install-cf")
>  > +(inputs
>  > + `(("m4" ,m4)
>  > +   ("perl" ,perl)))
>  > +(home-page "http://sendmail.org;)
>  > +(synopsis
>  > + "Highly configurable Mail Transfer Agent (MTA)")
>  > +(description
>  > + "Sendmail is a mail transfer agent (MTA) originally developed
>  > by Eric
>  > +Allman.  It is highly configurable and supports many delivery
>  > methods and many
>  > +transfer protocols.")
>  > +(license (non-copyleft "file://LICENSE"
>  > +   "See LICENSE in the distribution."
>  > +
>  
>  Thanks,
>  Alex

Re: [PATCH] gnu: Add sendmail

[John Darrington]

On Sat, Sep 17, 2016 at 10:50:35AM +0200, John Darrington wrote:
 On Sat, Sep 17, 2016 at 12:51:20PM +0800, Alex Vong wrote:
  
  > +"contrib/mmuegel" 
"devtools/bin/configure.sh")
  > +   (("/bin/sh") (which "bash")))
  > +
  > + (substitute* "devtools/bin/Build"
  > +   (("SHELL=/bin/sh") (string-append "SHELL=" (which 
"bash"
  > + #t))
  I think the `#t' is not neccessary here, since `substitute*' uses
  `substitute', which will either return #t or throw an exception.
 
 WTF??   Didn't you complain earlier this week when I *didn't* put #t in 
exactly this
 scenario??


Ahh no.  That was a different Alex.  I apologise.
  

J'


-- 
Avoid eavesdropping.  Send strong encrypted email.
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.



signature.asc
Description: Digital signature

Re: [PATCH] gnu: Add sendmail

[John Darrington]

On Sat, Sep 17, 2016 at 12:51:20PM +0800, Alex Vong wrote:
 Hello,
 
 Thanks for the patch. I do not know how to set up a mail server, so I
 can only comment on generic things. You will have to wait for sysadmin
 to help :)
 
 John Darrington  writes:
 
 > * gnu/packages/mail.scm (sendmail): New variable.
 > ---
 >  gnu/packages/mail.scm | 80 
+++
 >  1 file changed, 80 insertions(+)
 >
 > diff --git a/gnu/packages/mail.scm b/gnu/packages/mail.scm
 > index e344683..aff6a2c 100644
 > --- a/gnu/packages/mail.scm
 > +++ b/gnu/packages/mail.scm
 > @@ -1410,3 +1410,83 @@ provides HTML mail archiving with index, mail 
thread linking,
 >  etc; plus other capabilities including support for MIME and
 >  powerful user customization features.")
 >  (license gpl2+)))
 > +
 > +
 > +(define-public sendmail
 > +  (package
 > +(name "sendmail")
 > +(version "8.15.2")
 > +(source
 > + (origin
 > +   (method url-fetch)
 > +   (uri (string-append
 > + "ftp://ftp.sendmail.org/pub/sendmail/sendmail.;
 > + version ".tar.gz"))
 > +   (sha256
 > +(base32
 > + "0fdl9ndmspqspdlmghzxlaqk56j3yajk52d7jxcg21b7sxglpy94"
 > +(build-system gnu-build-system)
 > +(arguments
 > + `(#:tests? #f  ; There are no tests
 There is a test/ directory in the tarball which needs seperate
 compilation. The readme in the test/ directory says the test needs root
 privilege, but this is not a problem, since the build daemon has root
 privilege.

They need more than that.  They need the binary to be installed setuid.  So 
these
cannot be run as package tests.  They must be tested after the 
as-yet-to-be-written
service is installed.  But I'll update the comment to make it more clear.
 
 > +   #:phases
 > +   (modify-phases %standard-phases
 > + (add-before 'build 'replace-/bin/sh
 > +   (lambda _
 > + (substitute*
 > + (append
 > +  (list "smrsh/smrsh.c" "sendmail/conf.c" 
"contrib/mailprio"
 > +"contrib/mmuegel" "devtools/bin/configure.sh")
 > +  (find-files "." ".*\\.m4")
 > +  (find-files "." ".*\\.cf"))
 I think this can be simplified to:
 `("smrsh/smrsh.c" "sendmail/conf.c"
"contrib/mailprio" "contrib/mmuegel"
"devtools/bin/configure.sh"
.@(find-files "." ".*\\.m4")
,@(find-files "." ".*\\.cf"))
 using the quasi-quote quasi-unquote-splicing notation, which is similar
 to string interpolation in shell "foo bar $(CAR) $(TAR)".

Is that simpler?  It has more characters?
 
 > +"contrib/mmuegel" "devtools/bin/configure.sh")
 > +   (("/bin/sh") (which "bash")))
 > +
 > + (substitute* "devtools/bin/Build"
 > +   (("SHELL=/bin/sh") (string-append "SHELL=" (which 
"bash"
 > + #t))
 I think the `#t' is not neccessary here, since `substitute*' uses
 `substitute', which will either return #t or throw an exception.

WTF??   Didn't you complain earlier this week when I *didn't* put #t in exactly 
this
scenario??
 
 > + (replace 'configure
 > +   (lambda _
 > +
 > + ;; Render harmless any attempts to chown or chmod
 > + (substitute* "devtools/bin/install.sh"
 > +   (("owner=\\$2") "owner=''")
 > +   (("group=\\$2") "group=''"))
 > +
 > + (with-output-to-file "devtools/Site/site.config.m4"
 > +   (lambda ()
 > + (format #t "
 > +define(`confCC', `gcc')
 > +define(`confOPTIMIZE', `-g -O2')
 > +define(`confLIBS', `-lresolv')
 > +define(`confINSTALL', `~a/devtools/bin/install.sh')
 > +define(`confDEPEND_TYPE', `CC-M')
 > +define(`confINST_DEP', `')
 > +" (getcwd))
 > + (replace 'build
 > +   (lambda _
 > + (system* "sh" "Build")
 I think there is a missing `zero?' here.


Would that make any difference? Since only the last expression would get 
returned from lambda ?
 
 > + (with-directory-excursion "cf/cf"
 > +   (begin
 > + (copy-file "generic-linux.mc" "sendmail.mc")
 > + (zero? (system* "sh" "Build" "sendmail.cf"))
 > + (add-before 'install 'pre-install
 > +   (lambda _
 > + (let ((out (assoc-ref %outputs "out")))
 > +   (mkdir-p (string-append out "/usr/bin"))
 > +   (mkdir-p (string-append out "/usr/sbin"))
 > +   (mkdir-p (string-append 

Re: [PATCH] gnu: Add sendmail

[Alex Vong]

Hello,

Thanks for the patch. I do not know how to set up a mail server, so I
can only comment on generic things. You will have to wait for sysadmin
to help :)

John Darrington  writes:

> * gnu/packages/mail.scm (sendmail): New variable.
> ---
>  gnu/packages/mail.scm | 80 
> +++
>  1 file changed, 80 insertions(+)
>
> diff --git a/gnu/packages/mail.scm b/gnu/packages/mail.scm
> index e344683..aff6a2c 100644
> --- a/gnu/packages/mail.scm
> +++ b/gnu/packages/mail.scm
> @@ -1410,3 +1410,83 @@ provides HTML mail archiving with index, mail thread 
> linking,
>  etc; plus other capabilities including support for MIME and
>  powerful user customization features.")
>  (license gpl2+)))
> +
> +
> +(define-public sendmail
> +  (package
> +(name "sendmail")
> +(version "8.15.2")
> +(source
> + (origin
> +   (method url-fetch)
> +   (uri (string-append
> + "ftp://ftp.sendmail.org/pub/sendmail/sendmail.;
> + version ".tar.gz"))
> +   (sha256
> +(base32
> + "0fdl9ndmspqspdlmghzxlaqk56j3yajk52d7jxcg21b7sxglpy94"
> +(build-system gnu-build-system)
> +(arguments
> + `(#:tests? #f  ; There are no tests
There is a test/ directory in the tarball which needs seperate
compilation. The readme in the test/ directory says the test needs root
privilege, but this is not a problem, since the build daemon has root
privilege.

> +   #:phases
> +   (modify-phases %standard-phases
> + (add-before 'build 'replace-/bin/sh
> +   (lambda _
> + (substitute*
> + (append
> +  (list "smrsh/smrsh.c" "sendmail/conf.c" "contrib/mailprio"
> +"contrib/mmuegel" "devtools/bin/configure.sh")
> +  (find-files "." ".*\\.m4")
> +  (find-files "." ".*\\.cf"))
I think this can be simplified to:
`("smrsh/smrsh.c" "sendmail/conf.c"
   "contrib/mailprio" "contrib/mmuegel"
   "devtools/bin/configure.sh"
   .@(find-files "." ".*\\.m4")
   ,@(find-files "." ".*\\.cf"))
using the quasi-quote quasi-unquote-splicing notation, which is similar
to string interpolation in shell "foo bar $(CAR) $(TAR)".

> +"contrib/mmuegel" "devtools/bin/configure.sh")
> +   (("/bin/sh") (which "bash")))
> +
> + (substitute* "devtools/bin/Build"
> +   (("SHELL=/bin/sh") (string-append "SHELL=" (which "bash"
> + #t))
I think the `#t' is not neccessary here, since `substitute*' uses
`substitute', which will either return #t or throw an exception.

> + (replace 'configure
> +   (lambda _
> +
> + ;; Render harmless any attempts to chown or chmod
> + (substitute* "devtools/bin/install.sh"
> +   (("owner=\\$2") "owner=''")
> +   (("group=\\$2") "group=''"))
> +
> + (with-output-to-file "devtools/Site/site.config.m4"
> +   (lambda ()
> + (format #t "
> +define(`confCC', `gcc')
> +define(`confOPTIMIZE', `-g -O2')
> +define(`confLIBS', `-lresolv')
> +define(`confINSTALL', `~a/devtools/bin/install.sh')
> +define(`confDEPEND_TYPE', `CC-M')
> +define(`confINST_DEP', `')
> +" (getcwd))
> + (replace 'build
> +   (lambda _
> + (system* "sh" "Build")
I think there is a missing `zero?' here.

> + (with-directory-excursion "cf/cf"
> +   (begin
> + (copy-file "generic-linux.mc" "sendmail.mc")
> + (zero? (system* "sh" "Build" "sendmail.cf"))
> + (add-before 'install 'pre-install
> +   (lambda _
> + (let ((out (assoc-ref %outputs "out")))
> +   (mkdir-p (string-append out "/usr/bin"))
> +   (mkdir-p (string-append out "/usr/sbin"))
> +   (mkdir-p (string-append out "/etc/mail"))
> +   (setenv "DESTDIR" out)
> +   (with-directory-excursion "cf/cf"
> + (zero? (system* "sh" "Build" "install-cf")
> +(inputs
> + `(("m4" ,m4)
> +   ("perl" ,perl)))
> +(home-page "http://sendmail.org;)
> +(synopsis
> + "Highly configurable Mail Transfer Agent (MTA)")
> +(description
> + "Sendmail is a mail transfer agent (MTA) originally developed by Eric
> +Allman.  It is highly configurable and supports many delivery methods and 
> many
> +transfer protocols.")
> +(license (non-copyleft "file://LICENSE"
> +   "See LICENSE in the distribution."
> +

Thanks,
Alex